The real reason you sleep badly isn't your phone

August 5, 2026

If you are lying awake at two in the morning doing arithmetic on how many hours you have left, you have probably already tried the obvious things. Phone out of the room. Screens off at ten. Maybe the amber glasses.

And it mostly hasn't worked, because the thing that actually differs between you and someone who sleeps easily is not happening at night at all.

6.4 hours

Researchers strapped monitors to three groups living without electricity: the Hadza in Tanzania, the San in Namibia, the Tsimane in Bolivia.

They slept about 6.4 hours a night, which is less than you do. And almost none of them had insomnia. The reported figures are 1–2%, against something like 10–30% in industrialised populations, and in their languages there isn't a word for it. Treat those percentages as reported figures rather than precisely matched measurements, since the populations and definitions differ, but the direction is not in doubt.

Whatever they have, it isn't more sleep.

The evenings are almost the same

When one team measured evening light, from sunset to bedtime, on a week-long camping trip and then at home, the numbers came out close. About 8 lux out there. About 21 lux indoors. Almost the same darkness.

That's not a licence to leave every lamp on; we'll come back to your evening. But it does mean the evening is not where the two worlds diverge.

The days are not

Camping, people averaged around 4,500 lux across their waking hours. At home: 979. We live in permanent twilight.

A study in sunny San Diego found the average person got about 58 minutes a day of light bright enough to register. (It's an older figure, and "bright enough to count" is a threshold convention rather than a natural category.) Fifty-eight minutes, out of sixteen waking hours, in one of the sunniest cities in the United States.

A dark day makes the night worse

The two halves are connected. After a dim week, the same evening light shifts your body clock 60 to 70% harder than it does after a bright week. After a bright week, that evening light also suppresses less melatonin. The direction is well established even where the exact magnitude isn't.

Your morning decides how fragile your evening is. The same lamp, the same screen, at the same hour, does more damage to someone who spent the day indoors. That's why the evening-only fixes underperform: you have been trying to protect a system you spent all day making sensitive.

Two things you've heard that don't hold up

We are not sleeping less than a century ago. A review of 168 studies of objectively recorded sleep found no trend at all. The "we've lost one to two hours a night since our grandparents" line is one of the most repeated claims in this genre, and the recorded data doesn't support it.

Blue-light glasses are not established either. The Cochrane review found six sleep trials covering 148 people, with split results and very low certainty. That is not the same as "they don't work"; it means the evidence is too thin and too inconsistent to tell you either way, and the confident marketing has run far ahead of it.

You'll also notice what isn't in this article: nothing about sleep and Alzheimer's, nothing about inflammation, nothing recommending melatonin supplements or night mode. Those are the standard load-bearing claims of the sleep-alarm genre. They didn't survive the fact-check, so they're not here.

The daily contrast

Every ancestor you have woke to the same bright dawn and lost the same light at dusk, every single day, for hundreds of thousands of years. That daily contrast is the input your clock was built to read, and modern life has flattened it into a grey, undifferentiated smear. You are not bad at sleeping; your light is telling your clock the wrong time.

Regularity turns out to carry real weight too. When researchers tracked around 60,000 people, how regular their sleep was predicted how long they lived better than how much they slept. That's a predictive finding rather than a causal one, since nobody has shown that adopting a regular schedule extends your life, but it points at the same thing as everything else here.

The fix

Twenty minutes outside within an hour of waking. No sunglasses. Overcast is fine: an overcast morning is still around 10,000 lux, and your brightest room is around 500. You cannot get this indoors by a window and you cannot buy it in a bulb for less than the walk costs you.

Same time every day, because the regularity is the point. The shift builds with consistency; one heroic sunrise does very little.

At night, dim the room. Not warmer, dimmer. Brightness matters more than colour, which is why the instruction is a dimmer switch rather than an app setting. Get the day right and the evening becomes far more forgiving anyway.

That's the whole protocol. It costs twenty minutes and no money.

One caveat that matters: if you have chronic insomnia, months of it rather than a bad fortnight, the treatment with the best evidence is CBT-I, cognitive behavioural therapy for insomnia. A morning walk is a good habit for everyone and it is not a treatment for a clinical condition. Ask a doctor for the real thing.

Film version, with the sources on screen: The Real Reason You Sleep Badly Isn't Your Phone (2:31).